In this paper, a novel fault detection and isolation (FDI) framework based on kernel PCA (KPCA) and generalized likelihood ratio test (GLRT) that is capable of detecting and identifying faults is developed. Specifically, three main objectives are addressed. First, system model identification and residuals generation are addressed using KPCA model. Second, KPCA-based GLRT method is proposed to detect different types of faults in the systems. Third, partial KPCA (PKPCA)-based GLRT is developed for fault isolation. The proposed approach aims to apply a structured PKPCA -based GLRT to a set of sub-models. The fault detection and isolation performances using PKPCA-based GLRT are illustrated through two examples: a simulated continuous stirred tank reactor (CSTR) data and an air quality monitoring network data. The obtained results demonstrate the effectiveness of the partial KPCA-based GLRT method over the partial PCA-based GLRT method.
HarkatM.-F., MansouriM., NounouM. and NounouH., Enhanced data validation strategy of air quality monitoring network, Environmental Research160 (2018), 183–194.
2.
Castillo-ToledoB., Di GennaroS. and Anzurez-MarinJ., On the fault diagnosis problem for non-linear systems: A fuzzy sliding-mode observer approach 1, Journal of Intelligent & Fuzzy Systems20(4-5) (2009), 187–199.
3.
YeJ., Fault diagnoses of steam turbine using the exponential similarity measure of neutrosophic numbers, Journal of Intelligent & Fuzzy Systems30(4) (2016), 1927–1934.
4.
GeZ., SongZ. and GaoF., Review of recent research on databased process monitoring, Industrial & Engineering Chemistry Research52(10) (2013), 3543–3562.
5.
ShangJ., ChenM. and ZhangH., Fault detection based on augmented kernel mahalanobis distance for nonlinear dynamic processes, Computers & Chemical Engineering109 (2018), 311–321.
6.
ZhangY. and TengY., Process data modeling using modified kernel partial least squares, Chemical Engineering Science65(24) (2010), 6353–6361.
7.
AlcalaC.F. and QinS.J., Reconstruction-based contribution for process monitoring with kernel principal component analysis, Industrial & Engineering Chemistry Research49(17) (2010), 7849–7857.
8.
MorudT.E., Multivariate statistical process control; example from the chemical process industry, Journal of Chemometrics10(5-6) (1996), 669–675.
9.
WangK., ZhuP., XunP. and ShangB., An experimental study: An interpretative division method on principal component analysis, Journal of Intelligent & Fuzzy Systems33(1) (2017), 445–455.
10.
LeeJ.-M., QinS.J. and LeeI.-B., Fault detection and diagnosis based on modified independent component analysis, AIChE Journal52(10) (2006), 3501–3514.
11.
SchölkopfB., SmolaA. and MüllerK.-R., Nonlinear component analysis as a kernel eigenvalue problem, Neural Computation10(5) (1998), 1299–1319.
12.
ChoiS.W., LeeC., LeeJ.-M., ParkJ. H. and LeeI.-B., Fault detection and identification of nonlinear processes based on kernel pca, Chemometrics and Intelligent Laboratory Systems75(1) (2005), 55–67.
13.
DongD. and McAvoyT.J., Nonlinear principal component analysisâĂŤbased on principal curves and neural networks, Computers & Chemical Engineering20(1) (1996), 65–78.
14.
WangY., SunF. and LiX., Compound dimensionality reduction based multi-dynamic kernel principal component analysis monitoring method for batch process with large-scale data sets, Journal of Intelligent & Fuzzy Systems Preprint, 1–10.
15.
NaviM., MeskinN. and DavoodiM., Sensor fault detection and isolation of an industrial gas turbine using partial adaptive kpca, Journal of Process Control64 (2018), 37–48.
16.
SheriffM.Z., BotreC., MansouriM., NounouH., NounouM. and KarimM.N., Process monitoring using data-based fault detection techniques: Comparative studies,” in Fault Diagnosis and Detection. InTech, 2017.
17.
GertlerJ. and CaoJ., Design of optimal structured residuals from partial principal component models for fault diagnosis in linear systems, Journal of Process Control15(5) (2005), 585–603.
18.
GertlerJ., LiW., HuangY. and McAvoyT., Isolation enhanced principal component analysis, AIChE Journal45(2) (1999), 323–334.
19.
HuangY., GertlerJ. and McAvoyT.J., Sensor and actuator fault isolation by structured partial pca with nonlinear extensions, Journal of Process Control10(5) (2000), 459–469.
20.
RathiY., DambrevilleS. and TannenbaumA., Statistical shape analysis using kernel pca,” in Image processing: Algorithms and systems, neural networks, and machine learning, International Society for Optics and Photonics6064 (2006), p. 60641B..
21.
AversanoG., BellemansA., LiZ., CoussementA., GicquelO. and ParenteA., Application of reduced-order models based on pca & kriging for the development of digital twins of reacting flow applications, Computers & Chemical Engineering121 (2019), 422–441.
22.
XiaL., ChuJ. and GengZ., Process monitoring based on improved recursive pca methods by adaptive extracting principal components, Transactions of the Institute of Measurement and Control35(8) (2013), 1024–1045.
23.
MansouriM., BakloutiR., HarkatM.F., NounouM., NounouH. and HamidaA.B., Kernel generalized likelihood ratio test for fault detection of biological systems, IEEE Transactions on Nanobioscience17(4) (2018), 498–506.
24.
BotreC., MansouriM., NounouM., NounouH. and KarimM.N., Kernel pls-based glrt method for fault detection of chemical processes, Journal of Loss Prevention in the Process Industries43 (2016), 212–224.
25.
BanuU.S. and UmaG., Anfis based sensor fault detection for continuous stirred tank reactor, Applied Soft Computing11(2) (2011), 2618–2624.
